Clase 7

# sobre la EH19
rm(list=ls())
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

#install.packages("survey")
#install.packages("srvyr")
library(survey)

## Loading required package: grid

## Loading required package: Matrix

## Loading required package: survival

## 
## Attaching package: 'survey'

## The following object is masked from 'package:graphics':
## 
##     dotchart

library(srvyr)

## 
## Attaching package: 'srvyr'

## The following object is masked from 'package:stats':
## 
##     filter

#bases
load(url("https://github.com/AlvaroLimber/R_aru/raw/master/data/eh19.RData"))

Diseños muestrales en R (estimación)

Existen dos librerías

• survey: orientada al código R de fuente, interactúa con otras librarías como la convey. (No trabaja con dplyr)
• srvyr: Trabaja con la filosofía de dplyr

No es el mismo que STATA… [w=], lo mas cercano a Stata es el prefijo svy…

Estimador de Horvitz Thompson.

Parámetro del Total,

\[t_y=\sum_U{y_k}\]

El estimador,

\[\hat{t}_y=\sum_s \frac{y_k}{\pi_k}=\sum_s {y_k*\frac{1}{\pi_k}}=\sum_s {y_k*w_k}\]

En R…

#survey
sd1<-svydesign(ids = ~upm, strata=~estrato,weights =~factor ,data=eh19p)
svyhist(~aestudio,design = sd1)

hist(eh19p$aestudio)

svymean(~aestudio,design=sd1,na.rm=T)

##            mean     SE
## aestudio 8.1501 0.0722

t1<-svyby(~aestudio,~depto+area,design=sd1, svymean, na.rm=T,deff = T)
table(cv(t1)>0.10)

## 
## FALSE 
##    18

cv(t1)

## Chuquisaca.Urbana     La Paz.Urbana Cochabamba.Urbana      Oruro.Urbana 
##        0.04112228        0.01359558        0.01917086        0.03222194 
##     Potosí.Urbana     Tarija.Urbana Santa Cruz.Urbana       Beni.Urbana 
##        0.04172358        0.02540172        0.01862134        0.03162047 
##      Pando.Urbana  Chuquisaca.Rural      La Paz.Rural  Cochabamba.Rural 
##        0.03068992        0.07204062        0.03434233        0.03935627 
##       Oruro.Rural      Potosí.Rural      Tarija.Rural  Santa Cruz.Rural 
##        0.03638799        0.06865782        0.03523620        0.04566724 
##        Beni.Rural       Pando.Rural 
##        0.05773332        0.03930840

confint(t1)

##                      2.5 %    97.5 %
## Chuquisaca.Urbana 8.376470  9.845096
## La Paz.Urbana     9.568880 10.092802
## Cochabamba.Urbana 9.050918  9.757635
## Oruro.Urbana      9.146942 10.380153
## Potosí.Urbana     7.514075  8.852477
## Tarija.Urbana     8.377137  9.254977
## Santa Cruz.Urbana 8.646720  9.301789
## Beni.Urbana       7.669275  8.682688
## Pando.Urbana      8.362137  9.432504
## Chuquisaca.Rural  4.258291  5.658515
## La Paz.Rural      6.043890  6.916233
## Cochabamba.Rural  4.815068  5.619996
## Oruro.Rural       5.861075  6.761292
## Potosí.Rural      3.964398  5.197253
## Tarija.Rural      5.244833  6.023009
## Santa Cruz.Rural  5.044238  6.035987
## Beni.Rural        5.473158  6.869832
## Pando.Rural       5.940493  6.932248

summary(sd1)

## Stratified 1 - level Cluster Sampling design (with replacement)
## With (1047) clusters.
## svydesign(ids = ~upm, strata = ~estrato, weights = ~factor, data = eh19p)
## Probabilities:
##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
## 0.0004584 0.0029599 0.0042334 0.0051684 0.0060003 0.0418970 
## Stratum Sizes: 
##              11   12    13   14   21   22   23   24
## obs        4080 8769 10408 7491 2499 2554 2494 1310
## design.PSU   93  218   270  213   65   69   79   40
## actual.PSU   93  218   270  213   65   69   79   40
## Data variables:
##   [1] "folio"          "depto"          "area"           "nro"           
##   [5] "s02a_02"        "s02a_03"        "s02a_04a"       "s02a_04b"      
##   [9] "s02a_04c"       "s02a_05"        "s02a_06a"       "s02a_06b"      
##  [13] "s02a_06c"       "s02a_06d"       "s02a_06e"       "s02a_06_b"     
##  [17] "s02a_07_1"      "s02a_07_2"      "s02a_07_3"      "s02a_08"       
##  [21] "s02a_10"        "s03a_01a"       "s03a_01b"       "s03a_01c"      
##  [25] "s03a_01d"       "s03a_01d2_cod"  "s03a_01e"       "s03a_02"       
##  [29] "s03a_02e"       "s03a_03"        "s03a_03a"       "s03a_04"       
##  [33] "s03a_04npioc"   "s04a_01a"       "s04a_01b"       "s04a_01e"      
##  [37] "s04a_02a"       "s04a_02b"       "s04a_02e"       "s04a_03a"      
##  [41] "s04a_03b"       "s04a_03c"       "s04a_03d"       "s04a_03e"      
##  [45] "s04a_03f"       "s04a_03g"       "s04a_04a"       "s04a_04b"      
##  [49] "s04a_04e"       "S04A_0"         "S04A_1"         "S04A_2"        
##  [53] "s04a_05a"       "s04a_05b"       "s04a_05c"       "s04a_05d"      
##  [57] "s04a_05e"       "s04a_06a"       "s04a_07a"       "s04a_07a_e"    
##  [61] "s04a_06b"       "s04a_07b"       "s04a_07b_e"     "s04a_06c"      
##  [65] "s04a_07c"       "s04a_07c_e"     "s04a_06d"       "s04a_07d"      
##  [69] "s04a_07d_e"     "s04a_06e"       "s04a_07e"       "s04a_07e_e"    
##  [73] "s04a_06f"       "s04a_07f"       "s04a_07f_e"     "s04a_06g"      
##  [77] "s04a_07g"       "s04a_07g_e"     "s04a_08"        "s04a_08a1"     
##  [81] "s04a_08a2"      "s04a_08b"       "s04a_09"        "s04a_09a"      
##  [85] "s04b_11a"       "s04b_11b"       "s04b_12"        "s04b_13"       
##  [89] "s04b_14a"       "s04b_14b"       "s04b_15"        "s04b_15e"      
##  [93] "S04B_9"         "S04B_A"         "S04B_B"         "s04b_16"       
##  [97] "s04b_16e"       "S04B_6"         "S04B_7"         "S04B_8"        
## [101] "s04b_17"        "s04b_17e"       "S04B_3"         "S04B_4"        
## [105] "S04B_5"         "s04b_18"        "s04b_18e"       "S04B_0"        
## [109] "S04B_1"         "S04B_2"         "s04b_19"        "s04b_20a1"     
## [113] "s04b_20a2"      "s04b_20b"       "s04b_21a"       "s04b_21b"      
## [117] "s04b_21b2"      "s04c_22"        "s04c_23"        "s04d_24"       
## [121] "s04d_25"        "s04d_26"        "s04d_27a"       "s04d_27b"      
## [125] "s04e_28a"       "s04e_28b"       "s04e_29a"       "s04e_29b"      
## [129] "s04e_30a"       "s04e_30b"       "s04e_30c_cod"   "s04e_31a"      
## [133] "s04e_31b"       "s04e_31c"       "s04e_31d"       "s04e_31e"      
## [137] "s04e_31f"       "s04e_31_e"      "s04e_32a"       "s04e_32b"      
## [141] "s04e_33a"       "s04e_33b"       "s04_e_34a"      "s04f_34"       
## [145] "s04f_35a"       "s04f_35b"       "s04f_35c"       "s04f_35e"      
## [149] "s05a_01"        "s05a_01a"       "s05a_02a"       "s05a_02c"      
## [153] "s05a_03a"       "s05a_03c"       "s05a_04"        "s05a_05"       
## [157] "s05a_05_e"      "s05a_06a"       "s05a_06c"       "s05a_07a"      
## [161] "s05a_07b"       "s05a_08"        "s05a_09"        "s05b_10"       
## [165] "s05b_11"        "s05b_11_e"      "s05b_11a"       "s05c_13a"      
## [169] "s05c_13b"       "s05c_13c"       "s05c_13d"       "s05c_13e"      
## [173] "s05c_13f"       "s05c_13g"       "s05c_13h"       "s05c_13_e"     
## [177] "s05c_14a"       "s05c_14b"       "s05c_15a"       "s05c_15b"      
## [181] "s05d_17"        "s05d_18"        "s05d_19a"       "s05d_19b"      
## [185] "s05d_20a"       "s05d_20b"       "s05d_21a"       "s05d_21b"      
## [189] "s05d_21e"       "s05d_22a"       "s05d_22b"       "s05d_22c"      
## [193] "s05d_22d"       "s05d_22e"       "s05d_22f"       "s05d_22g"      
## [197] "s05d_22h"       "s05d_22i"       "s05d_22j"       "s05d_22k"      
## [201] "s05d_22l"       "s05d_22_e"      "s06a_01"        "s06a_02"       
## [205] "s06a_03"        "s06a_04"        "s06a_05"        "s06a_06aa"     
## [209] "s06a_06ab"      "s06a_06ac"      "s06a_06e"       "s06a_07"       
## [213] "s06a_08a"       "s06a_08b"       "s06a_09"        "s06a_09e"      
## [217] "s06a_10"        "s06a_10e"       "s06b_11a"       "s06b_11a_cod"  
## [221] "s06b_11b"       "s06b_12a"       "s06b_12a_cod"   "s06b_12b"      
## [225] "s06b_13"        "s06b_13a"       "s06b_13b"       "s06b_13c"      
## [229] "s06b_14"        "s06b_15aa"      "s06b_15ab"      "s06b_15ba"     
## [233] "s06b_15bb"      "s06b_15ca"      "s06b_15cb"      "s06b_15da"     
## [237] "s06b_15db"      "s06b_17"        "s06b_18"        "s06b_19a"      
## [241] "s06b_19b"       "s06b_20"        "s06b_20e"       "s06b_21a"      
## [245] "s06b_21b"       "s06b_22"        "s06b_23aa"      "s06b_23ab"     
## [249] "s06c_25a"       "s06c_25b"       "s06c_26a"       "s06c_26b"      
## [253] "s06c_27aa"      "s06c_27ab"      "s06c_27ba"      "s06c_27bb"     
## [257] "s06c_28a"       "s06c_28a1"      "s06c_28b"       "s06c_29a"      
## [261] "s06c_29b"       "s06c_30a"       "s06c_30a1"      "s06c_30a2"     
## [265] "s06c_30b"       "s06c_30b1"      "s06c_30b2"      "s06c_30c"      
## [269] "s06c_30c1"      "s06c_30c2"      "s06c_30d"       "s06c_30d1"     
## [273] "s06c_30d2"      "s06c_30e"       "s06c_30e1"      "s06c_30e2"     
## [277] "s06d_31a"       "s06d_31b"       "s06d_32aa"      "s06d_32ab"     
## [281] "s06d_32ba"      "s06d_32bb"      "s06d_32ca"      "s06d_32cb"     
## [285] "s06d_32da"      "s06d_32db"      "s06d_32ea"      "s06d_32eb"     
## [289] "s06d_32fa"      "s06d_32fb"      "s06d_32ga"      "s06d_32gb"     
## [293] "s06d_32ha"      "s06d_32hb"      "s06d_33a"       "s06d_33b"      
## [297] "s06d_34"        "s06e_35a"       "s06e_35a_cod"   "s06e_35b"      
## [301] "s06e_36"        "s06e_37"        "s06e_38a"       "s06e_38b"      
## [305] "s06e_39"        "s06e_40"        "s06e_40b"       "s06f_42a"      
## [309] "s06f_42b"       "s06f_43a"       "s06f_43a1"      "s06f_43b"      
## [313] "s06f_43b1"      "s06f_43c"       "s06f_43c1"      "s06f_44a"      
## [317] "s06f_44b"       "s06f_45aa"      "s06f_45ab"      "s06f_45ba"     
## [321] "s06f_45bb"      "s06f_45ca"      "s06f_45cb"      "s06f_45da"     
## [325] "s06f_45db"      "s06f_45ea"      "s06f_45eb"      "s06f_45fa"     
## [329] "s06f_45fb"      "s06f_45ga"      "s06f_45gb"      "s06f_45ha"     
## [333] "s06f_45hb"      "s06f_46a"       "s06f_46b"       "s06g_47"       
## [337] "s06g_48"        "s06g_49"        "s06g_49e"       "s06g_50"       
## [341] "s06g_50e"       "s06g_51"        "s06g_51e"       "s06g_52"       
## [345] "s06g_53"        "s06g_54"        "s06g_55"        "s07a_01a"      
## [349] "s07a_01b"       "s07a_01c"       "s07a_01d"       "s07a_01e"      
## [353] "s07a_01e0"      "s07a_01e1"      "s07a_01e1e"     "s07a_01e2"     
## [357] "s07a_01e2e"     "s07a_02a"       "s07a_02b"       "s07a_02c"      
## [361] "s07a_02ce"      "s07a_03a"       "s07a_03b"       "s07a_03c"      
## [365] "s07a_04a"       "s07a_04b"       "s07a_04c"       "s07a_04d"      
## [369] "s07b_05aa"      "s07b_05ab"      "s07b_05ba"      "s07b_05bb"     
## [373] "s07b_05ca"      "s07b_05cb"      "s07b_05da"      "s07b_05db"     
## [377] "s07b_05de"      "s07b_05ea"      "s07b_05eb"      "s07b_05ee"     
## [381] "s07c_06"        "s07c_07"        "s07c_08a"       "s07c_08b"      
## [385] "s07c_08e"       "s07c_09"        "s07c_09e"       "s07c_10"       
## [389] "s08a_01"        "s08a_03a"       "s08a_03b"       "s08a_03c"      
## [393] "s08a_03e"       "s08a_04"        "s08a_06"        "upm"           
## [397] "estrato"        "factor"         "tipohogar"      "cobersalud"    
## [401] "hnv_ult_a"      "quienatenparto" "dondeatenparto" "niv_ed"        
## [405] "niv_ed_g"       "cmasi"          "educ_prev"      "aestudio"      
## [409] "cob_op"         "caeb_op"        "pet"            "ocupado"       
## [413] "cesante"        "aspirante"      "desocupado"     "pea"           
## [417] "temporal"       "permanente"     "pei"            "condact"       
## [421] "phrs"           "shrs"           "tothrs"         "yprilab"       
## [425] "yseclab"        "ylab"           "ynolab"         "yper"          
## [429] "yhog"           "yhogpc"         "z"              "zext"          
## [433] "p0"             "p1"             "p2"             "pext0"         
## [437] "pext1"          "pext2"

hist(eh19p$factor)

summary(eh19p$factor)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   23.87  166.66  236.21  291.21  337.84 2181.61

quantile(eh19p$factor,c(0.05,0.1,0.9,0.95))

##        5%       10%       90%       95% 
##  78.91777 110.66443 514.98672 663.16821

#srvyr
sd2<-as_survey_design(sd1)
sd3<-eh19p %>% as_survey_design(ids=upm, strata=estrato, weights=factor)

sd2 %>% summarise(m_aes=survey_mean(aestudio,na.rm=T))

## # A tibble: 1 x 2
##   m_aes m_aes_se
##   <dbl>    <dbl>
## 1  8.15   0.0722

sd3 %>% summarise(m_aes=survey_mean(aestudio,na.rm=T))

## # A tibble: 1 x 2
##   m_aes m_aes_se
##   <dbl>    <dbl>
## 1  8.15   0.0722

sd3 %>% filter(s02a_03>=15) %>% summarise(m_aes=survey_mean(aestudio,na.rm=T))

## # A tibble: 1 x 2
##   m_aes m_aes_se
##   <dbl>    <dbl>
## 1  9.92   0.0800

t2<-sd3 %>% filter(s02a_03>=15) %>% group_by(depto,area) %>%  summarise(m_aes=survey_mean(aestudio,na.rm=T,deff=T,vartype=c("ci","cv","se")))

## Warning: The `add` argument of `group_by()` is deprecated as of dplyr 1.0.0.
## Please use the `.add` argument instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.

library(knitr)
kable(t2)

depto	area	m_aes	m_aes_low	m_aes_upp	m_aes_cv	m_aes_se	m_aes_deff
Chuquisaca	Urbana	10.895703	10.148803	11.642604	0.0349344	0.3806348	7.162531
Chuquisaca	Rural	5.693202	4.608445	6.777959	0.0971005	0.5528128	11.249735
La Paz	Urbana	11.595812	11.346048	11.845576	0.0109768	0.1272846	5.500503
La Paz	Rural	7.761397	7.192433	8.330360	0.0373585	0.2899545	2.984142
Cochabamba	Urbana	11.218832	10.845887	11.591777	0.0169412	0.1900601	7.237421
Cochabamba	Rural	6.114481	5.454174	6.774787	0.0550341	0.3365048	4.769871
Oruro	Urbana	11.637908	11.009817	12.265998	0.0275038	0.3200868	6.434192
Oruro	Rural	7.429932	6.775212	8.084653	0.0449073	0.3336581	3.888534
Potosí	Urbana	10.484944	9.787499	11.182389	0.0338992	0.3554312	4.338572
Potosí	Rural	5.216560	4.368462	6.064659	0.0828529	0.4322073	6.059638
Tarija	Urbana	10.821501	10.317403	11.325600	0.0237396	0.2568983	4.577131
Tarija	Rural	6.882555	6.059210	7.705900	0.0609646	0.4195923	5.004415
Santa Cruz	Urbana	11.053740	10.740448	11.367032	0.0144440	0.1596597	5.091124
Santa Cruz	Rural	6.695770	6.124629	7.266911	0.0434699	0.2910645	2.997842
Beni	Urbana	10.599824	10.070925	11.128722	0.0254284	0.2695368	4.254653
Beni	Rural	8.195720	7.419166	8.972275	0.0482870	0.3957469	3.680029
Pando	Urbana	11.428409	10.934781	11.922036	0.0220120	0.2515621	2.502778
Pando	Rural	8.783820	8.247287	9.320353	0.0311286	0.2734276	2.258157

Librería GGPLOT

## 
## Attaching package: 'plotly'

## The following object is masked from 'package:ggplot2':
## 
##     last_plot

## The following object is masked from 'package:stats':
## 
##     filter

## The following object is masked from 'package:graphics':
## 
##     layout

• Datos (base)
• aesthetics (x,y,col)
• Geometría (hist, boxplot, scatter, etc.)

ggplot(eh19p,aes(ylab))+geom_histogram()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

## Warning: Removed 23816 rows containing non-finite values (stat_bin).

ggplot(eh19p,aes(ylab))+geom_boxplot()

## Warning: Removed 23816 rows containing non-finite values (stat_boxplot).

ggplot(eh19p,aes(aestudio))+geom_bar()

## Warning: Removed 2903 rows containing non-finite values (stat_count).

ggplot(eh19p,aes(aestudio,weights=factor))+geom_bar()

## Warning: Removed 2903 rows containing non-finite values (stat_count).

ggplot(eh19p %>% filter(s02a_03>=15) ,aes(aestudio,weights=factor))+geom_bar()

## Warning: Removed 30 rows containing non-finite values (stat_count).

g1<-ggplot(eh19p %>% filter(s02a_03>=15) ,aes(aestudio,weights=factor))+geom_bar()

ggplotly(g1) 

## Warning: Removed 30 rows containing non-finite values (stat_count).